Deepfakes Take Brand Hijacking to New Levels — So What Will Brands Do?

This was a response to a media inquiry; I was a little slow to reply, but I liked what I said! (I'm kinda biased.)

One of the most common uses of deepfakes is old-fashioned brand hijacking to scam people. It's not surprising?as digital tools have taken this practice to new heights... or really lows. It's awfully funny that "brand safety" is a huge concern in digital advertising (i.e., don't place ads near content that might reflect poorly on the brand), but it doesn't concern scammers hijacking well-known brands.

Tom Hanks took to social media to address his persona being used to endorse products and The New York Times documented how AI impersonations?of Taylor Swift's voice were being used to push cooking ware. I find it incredibly?depressing seeing celebrities across the world (each global market has a few celebrities whose personas are continously hijacked)post to their followers, "Don't click on that, guys! I didn't endorse?that!"

This is part of the larger FizzCore/CelebCore digital advertising phishing attack that The Media Trust is constantly battling on the open web, but where scams hijacking brands and likenesses are rampant is social media.

There are limited audits of advertising and creatives before these social?media campaigns go live—often systems are self-serve and automated. I was served a ton of ads on X featuring Elon Musk's likeness hawking a non-existent piece of cryptocurrency (not a deepfake, though; The Media Trust had shut down the same campaign on the open web). Rarely have social media companies seen consequences?for helping empower scams with brand hijacking — the only case I can think of offhand is Martin Lewis suing Facebook in 2019.

Threat actors and scammers have of course embraced AI heartily — it's great for scaling campaigns,?making endless?assets, often hijacking brands. Major companies do some monitoring of brand abuse and report egregious examples to the Federal Trade Commission, or they press platforms, websites, and apps to shut down illicit campaigns. But they can't keep up with the scale of proliferation — It seems brand hijacking has just?become a cost of doing business.

The best way to fight AI is with AI — analysis of content at scale can reveal deepfakes as AI leaves signatures. Major brands (that includes celebrities) and companies need to redouble their efforts to shut down brand hijacking, particularly via deepfakes. The tools are out there, but it's a slog of a process. And as social media companies have apparently abdicated any strong notion of content moderation, it's questionable how responsive they will be in responding to reports... Particularly when there seem to be no consequences for letting scams and phishing run (and taking their cut for running the ad).??

The Media Trust does its part by identifying and neutralizing phishing and scam ads in real time on ad platforms and digital media publishers. But we've only scratched the surface of how much threat actors and scammers could harm consumers using deepfakes — in addition to the reputations of major brands. Brands need to be more proactive, not just for their own sake, but also to protect consumers.